WO2024186808A1 - Assessing blood oxygenation with pulse oximetry - Google Patents

Abstract

Provided are methods of assessing blood oxygenation of a subject with a pulse oximeter and treating the subject accordingly. After determining an approximate oxygen saturation with the pulse oximeter, this approximation can be further calibrated by considering both the skin pigmentation and perfusion index of the subject. Since traditional methods were found to have higher error rates when skin pigmentation is dark and perfusion index is low, the present methods can give more accurate determinations of oxygen saturation in such cases.

Description

ASSESSING BLOOD OXYGENATION WITH PULSE OXIMETRY

INTRODUCTION

[0001] Pulse oximeters are valuable diagnostic tools for estimating the oxygen saturation of blood in a patient. Pulse oximeters can include a light source that is placed on a body part of the patient, such as the finger, fingertip, toe, or earlobe. The pulse oximeter can emit light, such as both red and infrared light, that is directed towards the body part. Some of such emitted light will be absorbed by the body part, but some of such light will be transmitted through the body part. The intensity of such transmitted light is then recorded by a light sensor that is also part of the pulse oximeter. The amount of transmitted light is not recorded once, but instead the amount of transmitted light is recorded multiple times over a period of time. The graph of light intensity versus time is referred to as a photoplethysmogram (PPG) or simply as the “pleth”. Using the waveform of the PPG, it is possible to indirectly estimate blood oxygen saturation of hemoglobin in arterial blood, e.g. since highly oxygenated blood will absorb relatively more red light and relatively less infrared light than blood with lower oxygen saturation. Such pulse oximeter devices are well known in the field, such as those described by United States Patents 5,766, 127, 6,963,767, 6,912,413, 8,548,546, and 9,560,994.

[0002] However, even small errors in performance of pulse oximeters could lead to significant impacts on health and healthcare. Pulse oximeters cleared by the United States Food and Drug Administration (FDA) must read within 3% of arterial saturation (functional saturation, SaO ), but oximeter performance is degraded by the presence of numerous factors including skin melanin, dyshemoglobins, anemia, motion, and low perfusion (1,2).

[0003] Validation studies of pulse oximeter performance for International Organization for Standardization (ISO) or FDA clearance exclusively involve healthy young adult subjects with minimal co-morbidities that could interfere with device performance (3, 4). Since documentation of missed hypoxemia diagnosis due to melanin was documented by Bickler et al. and Feiner et al. in 2005 and 2007 (5, 6), the FDA has required inclusion of subjects with darkly pigmented skin in validation studies (7).

[0004] However, there is growing concern and evidence to suggest that current healthy human, laboratory-based study validation protocols may not adequately predict performance in real-world clinical settings (8). Recent retrospective studies report missed diagnoses of hypoxemia in patients with darkly pigmented skin at twice the rate compared to White patients (9-12). These findings have prompted inquiries by multiple regulatory bodies worldwide, including letters from Congress to the FDA and a new FDA communication (13, 14). Errors in pulse oximetry in hospitalized patients may lead to worse clinical outcomes and a perpetuation of disparities in health and healthcare (15, 16). Hence, the magnitude and cause of these errors in real world clinical settings remained uncertain.

[0005] References:

[0006] 1. Jubran A, Tobin MJ. Reliability of pulse oximetry in titrating supplemental oxygen therapy in ventilator-dependent patients. Chest. 1990;97(6):1420-1425.

[0007] 2. Severinghaus JW, Kelleher JF. Recent developments in pulse oximetry.

Anesthesiology. 1992;76(6):1018-1038.

[0008] 3. Batchelder PB, Raley DM. Maximizing the laboratory setting for testing devices and understanding statistical output in pulse oximetry. Anesth Analg. 2007;105(6 Suppl):S85-S94.

[0009] 4. Bickler PE, Feiner JR, Lipnick MS, Batchelder P, MacLeod DB, Severinghaus JW. Effects of Acute, Profound Hypoxia on Healthy Humans: Implications for Safety of Tests Evaluating Pulse Oximetry or Tissue Oximetry Performance. Anesth Analg. 2017 ; 124( 1 ) : 146-153. [0010] 5. Bickler PE, Feiner JR, Severinghaus JW. Effects of skin pigmentation on pulse oximeter accuracy at low saturation. Anesthesiology. 2005 ; 102(4):715-719.

[0011] 6. Feiner JR, Severinghaus JW, Bickler PE. Dark skin decreases the accuracy of pulse oximeters at low oxygen saturation: the effects of oximeter probe type and gender. Anesth Analg. 2007;105(6 Suppl):S18-S23.

[0012] 7. Pulse Oximeters- Premarket Notifications Submissions [510(k)s] Guidance for Industry and Food and Drug Administration Staff. In: FDA; 2007.

[0013] 8. Okunlola OE, Lipnick MS, Batchelder PB, Bernstein M, Feiner JR, Bickler PE. Pulse

Oximeter Performance, Racial Inequity, and the Work Ahead. Respir Care. 2022;67(2):252-257. [0014] 9. Andrist E, Nuppnau M, Barbaro RP, Valley TS, Sjoding MW. Association of Race With Pulse Oximetry Accuracy in Hospitalized Children. JAMA Netw Open. 2022;5(3):e224584. [0015] 10. Burnett GW, Stannard B, Wax DB, et al. Self-reported Race/Ethnicity and

Intraoperative Occult Hypoxemia: A Retrospective Cohort Study. Anesthesiology. 2022; 136(5):688-696.

[0016] 11. Sjoding MW, Dickson RP, Iwashyna TJ, Gay SE, Valley TS. Racial Bias in Pulse

Oximetry Measurement. V Engl J Med. 2020;383(25):2477-2478.

[0017] 12. Valbuena VSM, Merchant RM, Hough CL. Racial and Ethnic Bias in Pulse

Oximetry and Clinical Outcomes. JAMA Intern Med. 2022;182(7):699-700.

[0018] 13. Pulse Oximeter Accuracy and Limitations: FDA Safety Communication. In:2021.

[0019] 14. Warren E, Wyden R, Booker CA. Letter to the United States Food and Drug

Administration (FDA) regarding accuracy of pulse oximeters across racially diverse patients, 25 January 2021, www.warren.senate.gov/imo/media/doc/2020.01.25%20Letter%20to%20FD A%20re%20Bias% 20in%20Pulse%200ximetry%20Measurements.pdf, accessed 9 February 2023.

[0020] 15. Henry NR, Hanson AC, Schulte PJ, et al. Disparities in Hypoxemia Detection by

Pulse Oximetry Across Self-Identified Racial Groups and Associations With Clinical Outcomes. Crit Care Med. 2022;50(2):204-211.

[0021] 16. Wong A- KI, Charpignon M, Kim H, et al. Analysis of Discrepancies Between Pulse

Oximetry and Arterial Oxygen Saturation Measurements by Race and Ethnicity and Association With Organ Dysfunction and Mortality. JAMA Network Open. 2021;4(ll):e2131674.

SUMMARY

[0022] Provided are methods of assessing blood oxygenation of a subject with a pulse oximeter and treating the subject accordingly. After determining an approximate oxygen saturation with the pulse oximeter, this approximation can be further improved by considering both the skin pigmentation and perfusion index of the subject. Since traditional methods were found to have higher error rates when skin pigmentation is dark and perfusion index is low, the present methods can give more accurate determinations of oxygen saturation in such cases.

BRIEF DESCRIPTION OF THE DRAWINGS

[0023] FIG. 1 shows a flowchart for practicing methods of assessing blood oxygenation of a subject and treating the subject accordingly.

[0024] FIG. 2 shows a flowchart for determining raw SpO2.

[0025] FIG. 3 shows a flowchart for determining skin pigmentation with pulse oximeter data.

[0026] FIG. 4 shows determination of perfusion index.

[0027] FIG. 5 shows a method of generating adjusted SpO₂.

[0028] FIG. 6 shows a method of generating confidence intervals for the adjusted SpO₂.

[0029] FIG. 7 shows demographic data for a blood oxygenation study.

[0030] FIG. 8 shows SpO₂ compared with skin pigmentation and other variables.

[0031] FIG. 9A shows pulse oximeter bias for different perfusion index levels.

[0032] FIG. 9B shows pulse oximeter bias for different skin pigmentations.

[0033] FIG. 9C shows pulse oximeter bias for different ranges of steady-state hypoxemia.

[0034] FIG. 10A shows bias as a function of SaO₂ and skin pigment for Masimo devices.

[0035] FIG. 10B shows bias as a function of SaO₂ and skin pigment for Nellcor devices.

[0036] FIG. 11 A shows bias as a function of perfusion and skin pigment for Masimo devices.

[0037] FIG. 1 IB shows bias as a function of perfusion and skin pigment for Nellcor devices.

[0038] FIG. 12 shows aggregated statistics regarding bias under different conditions. [0039] FIG. 13 shows percentage incidences of missed hypoxemia diagnoses.

[0040] FIG. 14 shows statistical data regarding blood oxygenation bias.

[0041] FIG. 15 shows blood oxygenation data for Masimo and Nellcor devices.

DETAILED DESCRIPTION

[0042] Provided are methods of assessing blood oxygenation of a subject with a pulse oximeter and treating the subject accordingly. After determining an approximate oxygen saturation with the pulse oximeter, this approximation can be further calibrated by considering both the skin pigmentation and perfusion index of the subject. Since traditional methods were found to have higher error rates when skin pigmentation is dark and perfusion index is low, the present methods can give more accurate determinations of oxygen saturation in such cases.

[0043] Before the present invention is described in greater detail, it is to be understood that this invention is not limited to particular embodiments described, as such may vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting, since the scope of the present invention will be limited only by the appended claims.

[0044] Where a range of values is provided, it is understood that each intervening value, to the tenth of the unit of the lower limit unless the context clearly dictates otherwise, between the upper and lower limits of that range is also specifically disclosed. Each smaller range between any stated value or intervening value in a stated range and any other stated or intervening value in that stated range is encompassed within the invention. The upper and lower limits of these smaller ranges may independently be included or excluded in the range, and each range where either, neither or both limits are included in the smaller ranges is also encompassed within the invention, subject to any specifically excluded limit in the stated range. Where the stated range includes one or both of the limits, ranges excluding either or both of those included limits are also included in the invention.

[0045] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, some potential and exemplary methods and materials may now be described. Any and all publications mentioned herein are incorporated herein by reference to disclose and describe the methods and/or materials in connection with which the publications are cited. It is understood that the present disclosure supersedes any disclosure of an incorporated publication to the extent there is a contradiction. [0046] It must be noted that as used herein and in the appended claims, the singular forms "a", "an", and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to "a droplet" includes a plurality of such droplets and reference to "the discrete entity" includes reference to one or more discrete entities, and so forth. It is further noted that the claims may be drafted to exclude any element, e.g., any optional element. As such, this statement is intended to serve as antecedent basis for use of such exclusive terminology as “solely”, “only” and the like in connection with the recitation of claim elements, or the use of a “negative” limitation.

[0047] The publications discussed herein are provided solely for their disclosure prior to the filing date of the present application. Further, the dates of publication provided may be different from the actual publication dates which may need to be independently confirmed. To the extent the definition or usage of any term herein conflicts with a definition or usage of a term in an application or reference incorporated by reference herein, the instant application shall control.

[0048] As will be apparent to those of skill in the art upon reading this disclosure, each of the individual embodiments described and illustrated herein has discrete components and features which may be readily separated from or combined with the features of any of the other several embodiments without departing from the scope or spirit of the present invention. Any recited method can be carried out in the order of events recited or in any other order which is logically possible.

METHODS

[0049] Provided are methods of assessing blood oxygenation of a subject with a pulse oximeter. Such methods can also include treating the subject based on the assessed blood oxygenation. An embodiment of such methods is shown in FIG. 1.

[0050] For instance, in some cases the method includes: recording data from a pulse oximeter positioned on a body part of the subject, wherein the pulse oximeter data comprises the amount of light emitted from the pulse oximeter that is transmitted through the body part over time; determining a raw peripheral oxygen saturation (raw SpOp and a perfusion index based on the pulse oximeter data; determining skin pigmentation of the subject; adjusting the raw SpCh based on the perfusion index and the skin pigmentation of the subject, thereby generating an adjusted SpCh and a confidence interval for the adjusted SpOz; determining a high probability of hypoxemia based on the adjusted SpCh, the confidence interval, or a combination thereof; and treating the subject for the high probability of hypoxemia.

[0051] As discussed above, the method includes recording data from a pulse oximeter positioned on a body part of the subject. In some cases the body part is a finger or ear lobe. However, any suitable body part can be employed.

[0052] Next, the pulse oximeter data is used to determine a raw peripheral oxygen saturation (raw SpO₂) and a perfusion index. Any suitable method of determining raw SpO₂ can be used, wherein such methods are well known in the art, such as in United States Patents 5,766,127, 6,963,767, 6,912,413, 8,548,546, and 9,560,994, which are incorporated herein by reference.

[0053] For example, as shown in FIG. 2, in some cases the raw SpO₂ is calculated from the AC and DC component of the graph of light intensity versus time (i.e. photoplethysmogram or PPG). AC refers to the peak-to-peak amplitude of transmitted light over time. Stated in another manner, AC refers to the difference in light transmission value between the maximum transmission and the minimum transmission during a particular period of time. In addition, DC refers to the lowest light transmission value during a particular period of time. Stated in another manner, DC refers to the lowest transmission value minus 0% transmission, which is equal to the lowest transmission value. Such AC and DC concepts are shown visually in FIG. 2. The AC and DC calculations are performed for a first wavelength of light, e.g. red light. In some cases, the red light comprises light at 600 nm. The AC and DC calculations are repeated with a second wavelength of light, e.g. infrared light, such as infrared light that comprises light at 900 nm.

[0054] Next, the AC and DC values at both wavelengths can be used in an algorithm to calculate the raw SpO₂. For instance, a value called R can be calculated based on the equation below. AC first) - DC (first)

AC (second) - DC (second)

[0055] This current value of R can be compared to previous calibrations in order to determine the raw SpO₂ of the current patient. For instance, previous calibration studies could have been used to calculate certain R values for a patient while blood was simultaneously withdrawn from the patient and subjected to an arterial blood gas (ABG) test for determining arterial saturation (SaO₂). Such previous calibration studies could have been performed with many patients under many different conditions, thereby generating a calibration table between R value and SaO₂ determined by ABG. Therefore, the R value of the current patient can be compared to such calibration tables to determine raw SpO₂.

[0056] The pulse oximeter data is also used to determine perfusion index. Methods of determining the perfusion index are known in the art, such as in United States Patents 5,766,127 and 6,912,413, along with US Patent Application Publication 2008/188728, which are incorporated herein by reference. In some cases, perfusion index is calculated based on the AC and DC components of the photoplethysmogram (PPG), as shown in FIG. 4. In some cases, the perfusion index is AC divided by DC.

[0057] The method of assessing blood oxygenation also includes the step of determining skin pigmentation of a subject. Such pigmentation can affect the amount and wavelength of light absorbed by the skin, and therefore skin pigmentation is an important variable to calibrate for when generating the adjusted SpO2. For instance, the subject can be classified as having light skin pigmentation, medium skin pigmentation, or dark skin pigmentation.

[0058] In some embodiments, the skin pigmentation is determined by visual observation of the skin of the patient, e.g. based on the Fitzpatrick scale (e.g. Fitzpatrick, "Ultraviolet-induced pigmentary changes: Benefits and hazards", Therapeutic Photomedicine, Karger, vol. 15 of "Current Problems in Dermatology", 1986, pp. 25-38). The skin pigmentation can also be determined by obtaining skin pigmentation information from a medical database, e.g. wherein skin pigmentation was determined previously.

[0059] However, visual assessment of skin pigmentation can introduce errors, such as user error. Furthermore, the skin might have a different color on one body part than another body part. For example, in many dark skin individuals, the skin on the inside of the fingers is significantly lighter than skin elsewhere on the body.

[0060] Thus, to avoid such errors, in some cases the skin pigmentation is determined from the pulse oximeter data. For instance, pulse oximeter measured light signals from the absorption of red and infrared light can be used. Using this light signal data, a correction individualized for each patient or skin type can be made. For instance, as shown in FIG. 3, a previous calibration of historical data from multiple patients can be used to generate a calibration table. The calibration table can include the ratio of transmitted visible (e.g. red) light to transmitted infrared light along with the skin pigmentation determined by another method. This other method can employ quantitative and laboratory methods of determining skin pigmentation accurately.

[0061] Therefore, for the current patient, the pulse oximeter can measure the ratio of transmittances for visible (e.g. red) and infrared light and compare that ratio to the historical data. Hence, using this comparison with historical data allows skin pigmentation to be determined instantaneously without the effort or potential errors of human visual observation.

[0062] The method of assessing blood oxygenation also includes adjusting the raw SpO based on the perfusion index and the skin pigmentation of the subject. This adjusting step generates both an adjusted SpCh and a confidence interval based for the SpO2. In some cases, the confidence interval is 1.96 standard deviations. For instance, the raw SpCh could be 96% and the adjusted SpO₂ could be 92% with a confidence interval of from 90% to 94% (1.96 o). In another example, the raw SpO2 could be 93% and the adjusted SpO2 could be 85% with a confidence interval of from 79% to 91% (1.96 o). A confidence interval of 1.96 standard deviations can also be referred to as the 95% confidence interval. The adjusted SpO can be used to more accurately estimate the true blood oxygenation of the patient than the raw SpOr. Furthermore, the size of the confidence interval can help provide warning of low oxygen saturation, e.g. an 85% estimated SpC might not require medical intervention, but the large confidence interval generated a lower confidence limit of 79%, which could require medical attention.

[0063] In some cases, the adjusting is performed while considering perfusion index independently from skin pigmentation. For example, the raw SpCh can be adjusted based on perfusion index to generate a partially-adjusted SpC , and then the partially-adjusted SpCh is adjusted based on skin pigmentation to generate the adjusted SpCh.

[0064] However, in other cases, the adjusting is performed while considering the perfusion index and skin pigmentation together. Such adjusting can in some cases generate more accurate SpCh values. For example, lower perfusion in light skin people might generate an average error of 1% between raw SpO and arterial saturation (SaCh). Also, dark skin in people with normal perfusion might cause an average error of 2%. However, measurements for patients with both dark skin and low perfusion might have an average error of 6%, which is significantly higher than the estimated 3% error when considering perfusion and skin pigmentation separately. Similarly, there can be larger standard deviations for patients with both dark skin and low perfusion. Hence, considering perfusion and skin pigmentation simultaneously can also help by correctly generating the large confidence intervals for such patients that are warranted.

[0065] FIG. 5 shows an embodiment of generating an adjusted SpO2. In FIG. 5, all skin pigmentations are described as belonging to one of three categories: light, medium, and dark. Additionally, all perfusion indexes are sorted into three categories: less than 1, 1 to 2, or 2 or more. Furthermore, the raw SpOr is adjusted based on linear regression using a y = mx + b type equation. Different coefficients “m” and “b” were generated for each of the 9 possible combinations of skin pigmentation and perfusion index. For the current patient, the appropriate “m” and “b” values used in the y = mx + b type equation, thereby allowing for the calculation of the adjusted SpCb. For confidence intervals, each of the 9 categories could have a different standard deviation (o). In some cases, the adjusting is performed based on a linear relationship between raw SpO2 and adjusted SpO2, i.e., the adjusting is performed using an equation of the y = mx + b type.

[0066] Generating the adjusted SpO2 and confidence intervals can be based on historical data from multiple individuals. In particular, the historical calibration tests can simultaneously measure skin pigmentation, perfusion index, raw SpO2, and arterial oxygen saturation (SaO2) through an arterial blood gas (ABG) test. After the collection of data for many individuals under many different variables, such historical data can be used to estimate the average error between raw SpO₂ and SaO₂ for certain combinations of skin pigmentation and perfusion index. Furthermore, the standard deviation of the errors between SpO₂ and SaO₂ can be calculated. Based on these historical average errors and historical standard deviations, the raw SpO₂ of the current patient can be used to determine the adjusted SpO₂ and confidence interval.

[0067] The method of assessing blood oxygenation also includes determining a high probability of hypoxemia. For instance, if the adjusted SpO₂ is below a predetermined threshold, such as below 90%, the patient can be determined to have an unacceptably high probability of hypoxemia. In some cases, the adjusted SpO₂ is above the predetermined threshold, but the lower range of the confidence interval is below a certain threshold (e.g. below 90%), and therefore a high probability of hypoxemia is also determined. In some cases, the adjusted SpO₂ is below 85%, such as below 80%.

[0068] The method of assessing blood oxygenation also includes treating the subject for the high probability of hypoxemia. For instance, the blood oxygenation could be assessed through an arterial blood gas (ABG) test. The treating can also include administering oxy gen-enriched air to the subject, administering a bronchodilator to the subject, administering a diuretic to the subject, or a combination thereof. Any other suitable therapy medically indicated for high probability of hypoxemia can also be used.

[0069] Also provided by the present disclosure are methods of determining skin pigmentation of a body part of a subject. In some cases, the determination is performed with a pulse oximeter, and in other cases the determination is performed with a device that is not a pulse oximeter. In some cases, the method includes the steps of: recording an amount of visible light emitted from a light source that is transmitted through the body part; recording an amount of infrared light emitted from the light source that is transmitted through the body part; and comparing the ratio of transmitted visible light to transmitted infrared light and thereby determining skin pigmentation at the body part.

[0070] In some cases, the visible light is red light, such as red light comprising light at 660 nm. In some cases the infrared light comprises light at 900 nm. In some cases, the body part is a finger or an ear lobe.

DEVICES

[0071] Provided by the present disclosure are pulse oximeters for assessing blood oxygenation in a subject and communicating the results of the assessment. In some cases, the pulse oximeter is configured to: record pulse oximeter data from a body part of the subject, wherein the pulse oximeter data comprises the amount of light emitted from the pulse oximeter that is transmitted through the body part over time; determine a raw peripheral oxygen saturation (raw SpO₂) and a perfusion index based on the pulse oximeter data; determine skin pigmentation of the subject based on the pulse oximeter data or receive skin pigmentation data; adjust the raw SpO₂ based on the perfusion index and the skin pigmentation of the subject, thereby generating an adjusted SpO₂ and a confidence interval for the adjusted SpO₂; and communicating the adjusted SpO₂ and confidence interval.

[0072] In some cases, the communicating includes providing a visual indication of the adjusted SpO₂ and confidence interval. For instance, the visual indication can be provided by a visual display that is a part of the pulse oximeter. The visual display can numerically show the adjusted SpO₂ value, the lower limit of the confidence interval, and the upper limit of the confidence interval. In some cases, pulse oximeter is configured to electronically transmit the adjusted SpO₂ and the confidence interval to another device such that the other device can visually display the adjusted SpO₂ and confidence interval. In some cases, communicating the adjusted SpO₂ and confidence interval includes providing an auditory alert that the adjusted SpO₂ or confidence interval is outside a predetermined range or threshold, e.g. as described above.

[0073] The pulse oximeter is configured to record pulse oximeter data, determine raw SpO₂, and determine perfusion index in the same manner discussed above regarding the method of determining oxygen saturation. In addition, in some cases the pulse oximeter is configured to determine skin pigmentation from the pulse oximeter data, e.g. as discussed above. In some cases, the pulse oximeter is configured to receive skin pigmentation data, i.e. to receive the skin pigmentation. For instance, the pulse oximeter can receive the skin pigmentation from a medical database or from a user who selected a skin pigmentation on a separate device that transmits the skin pigmentation to the pulse oximeter. The pulse oximeter can also determine the adjusted SpO₂ and confidence interval in the manners discussed above regarding the method of determining oxygen saturation.

[0074] Also provided by the present disclosure is a controller for determining skin pigmentation of a body part of the subject. In some cases, the controller is configured to: recording an amount of visible light emitted from a light source that is transmitted through the body part; recording an amount of infrared light emitted from the light source that is transmitted through the body part; and comparing the ratio of transmitted visible light to transmitted infrared light and thereby determining skin pigmentation at the body part.

[0075] In some cases, the body part is a finger or an ear lobe. In some cases, the light source is pulse oximeter, e.g. wherein the pulse oximeter has a light emitting element that emits the light. In some cases, the visible light is red light, e.g. 660 nm light. In some cases, the infrared light comprises light at 900 nm.

KITS

[0076] Also provided are kits that include devices for performing the methods described herein. In some cases, the kit is a kit for assessing blood oxygenation in a subject and communicating the results of the assessment. For instance, the system can include: a pulse oximeter as discussed above; and packaging containing the pulse oximeter.

[0077] In some cases, the kit further includes instructions for using the pulse oximeter, e.g. that describe a method of determining oxygen saturation with the pulse oximeter. In some embodiments the kit also includes wires for connecting the pulse oximeter to an electrical supply. The kit can also includes wires for connecting the pulse oximeter to another electronic device, such as a computer or a device for providing an auditory alert.

EXAMPLES

[0078] The following examples are put forth so as to provide those of ordinary skill in the art with a complete disclosure and description of how to make and use the present invention, and are not intended to limit the scope of what the inventors regard as their invention nor are they intended to represent that the experiments below are all or the only experiments performed. Efforts have been made to ensure accuracy with respect to numbers used but some experimental errors and deviations should be accounted for.

Example 1

[0079] A purpose of the study was to quantitatively test the hypothesis that pulse oximeter SpO overestimates SaCh more frequently in the presence of increased skin melanin and low perfusion than in the presence of either condition alone. The study was designed to avoid potential confounding factors including mistiming of blood samples and oximeter readings, inconsistent use of functional versus fractional saturation, and self-reported race used as a surrogate for skin color [0080] The blood oxygenation of 146 health subjects was measured. Of the 146 subjects, 25 had light skin (Fitzpatrick class I-II), 78 had medium skin (classes III-IV), and 43 had dark skin (classes V-VI).

[0081] Blood oxygenation was recorded by both pulse oximeter readings (SpOz) and arterial oxygen saturation (hemoximetry SaO2) during stable hypoxemia (SaOz 68%- 100%). Specifically, 9,763 recordings of matched SpO2 and SaO2 were measured. Two different pulse oximeters were used: the Nellcor N-595™ and the Masimo Radical 7™, which are in prevalent use in North America, Europe, and the Asia-Pacific region (Pulse Oximeter Market by Product Global Forecast to 2027. 2022, www.marketsandmarkets.com/Market-Reports/pulse-oximeter-market-

68168578.html, accessed 30 September 2022).

[0082] Perfusion index (PI) was measured as percent infrared light modulation by the pulse detected by the pulse oximeter probe, with low perfusion categorized as PI < 1%.

[0083] “Pulse oximeter bias” was quantified as the difference between the SaO2 and SpO2 measurements from the same subject performed at the same time.

[0084] Methods

[0085] Perfusion Index

[0086] The perfusion index (PI) from each oximeter probe was recorded continuously. The Nellcor PI was divided by 10 to make the values numerically comparable to Masimo. Since no standard exists to define low perfusion conditions for pulse oximetry, a sensitivity analysis comparing bias and root mean square error (Arms) at various levels of PI was performed. Mean bias in oximeter readings increased approximately 5 -fold in grouping of readings with PI >2 versus PI<0.5 %. Scatterplots revealed that the relationship of bias to PI becomes very small at high values (PI^ 2%) in a non-linear fashion. For groups of data with Pl values of 1.0-1.5 %, 0.5- 1.0 % and <0.5%, the P value of the relationship to mean bias or median absolute bias was <0.00001. Accordingly, low perfusion was defined for the purpose of this study as PI< 1.0%.

[0087] Hypoxemia Protocol

[0088] Participants were in a 30-degree head up semi-recumbent position for the study. After local anesthesia, a 22-ga catheter was placed in a radial artery. To create steady-state hypoxemia, subjects breathed controlled air-nitrogen-CO2 mixtures via a mouthpiece and partial re-breathing system to achieve multiple stable plateaus of arterial oxygenation and a relatively constant endtidal CO2 per ISO and FDA guidelines (Premarket Notifications Submissions [510(k)sJ Guidance for Industry and Food and Drug Administration Staff. FDA 2013; Guideline for Evaluating and documenting SpO2 Accuracy in human Subjects. International Standards Organization; 2017). A more detailed description of testing the accuracy of pulse oximetry devices in the laboratory setting has been published (Batchelder PB, Raley DM. Maximizing the laboratory setting for testing devices and understanding statistical output in pulse oximetry. Anesth Analg. 2007; 105(6 Suppl) :S 85- S94; Bickler PE, Feiner JR, Lipnick MS, Batchelder P, MacLeod DB, Severinghaus JW. Effects of Acute, Profound Hypoxia on Healthy Humans: Implications for Safety of Tests Evaluating Pulse Oximetry or Tissue Oximetry Performance. Anesth Analg. 2017;124(l): 146- 153).

[0089] Warming pads on the hands and/or forearms were used on some subjects to improve perfusion, but no particular level of perfusion was sought, and subjects with low perfusion were not removed from the study. Blood samples for SaO2 measurement (Radiometer ABL90 Flex, Copenhagen, Denmark) were collected through the radial artery catheter.

[0090] A Masimo Radical 7 (Non-touch screen model, Masimo Inc., Irvine, CA) with a DCI clip-type adult reusable finger sensor or ear clip sensor, and Nellcor N-595 (Medtronic Inc, Minneapolis, MN) with a reusable finger clip sensor (Nellcor DS-100A) were placed on one of the middle three fingers or ear. Probes were re-positioned as necessary to ensure proper placement throughout the study. No subjects wore nail polish. Data from the pulse oximeters were recorded at 2 Hz from the instruments’ serial output ports using a computer running Lab VIEW 15.0. (National Instruments, Austin, Texas). Data from the Masimo were transmitted at 1 Hz, while the data from the Nellcor were transmitted at 0.5 Hz. Recorded data included SpO2, heart rate and perfusion index from each device.

[0091] Stable SaO2 plateaus between 60% and 100% were targeted by the study physician who adjusted the inspired gas mixture. At each level, two arterial blood samples were collected, approximately 30 seconds apart, each during steady- state conditions. Some (233 of 9763) SpO2 values were eliminated for obvious oximeter signal dropout, or failure to reach an appropriate stable plateau. Only samples with stable SpO2 readings (i.e., fluctuation <2% per minute) were included in the analysis. No subjects had MetHb or COHb outside the normal range as measured by the hemoximeter (Radiometer ABL90 Flex, Copenhagen, Denmark).

[0092] Statistics

[0093] Bias (or error) was computed as SpO2 minus the corresponding arterial blood SaO2. Bias assessment parameters included the mean bias ( “ accuracy ” ), standard deviation of the bias ( “precision ” ), and root mean square error (Arms) — an overall performance measure used by the FDA. Arms was calculated as the square root of the mean difference between SpO2 and SaO2, squared. The 95% confidence limits of Arms were determined using bootstrapping (random resampling with replacement) with 50,000 repetitions. Prior sensitivity analyses showed that this was sufficient, as the results did not change at the reported level of significance. The 95% limits of agreement (LOA) were calculated as 1.96 *SD according to Bland and Altman agreement analysis with adjustments for multiple measurements for each individual according to the “ Method Where the True Value Varies ” ( Bland JM AD. Statistical Methods for Assessing Agreement Between Two Methods of Clinical Measurements. The Lancet. 1986;327(8476):307- 310). The 95% confidence limits for the LOA were also determined using bootstrapping.

[0094] The primary analysis was a multivariable mixed-effects model of bias that accounts for repeated-measures changes in individuals at different SaO2 levels and different perfusion index levels, with fixed effects for gender and skin classification. Three skin pigment categories (light medium and dark) were used in the model for clarity, but results were confirmed with the six Fitzpatrick skin classifications. A three-level perfusion index scale was used: PI<1%, PI ^ 1% and <2%, and PI ^2%. A complete model with all terms, and including all possible interactions of skin pigmentation, PI and SaO2, was performed to confirm the results. Analysis was performed in both MatLab (The Mathworks, Inc., Boston, MA) and Stata 17.0 (Statacorp., College Station, TX).

[0095] Given the possibility of confounding, possible interactions between the variables in the model were also analyzed. Interactions between PI and both sex and three skin categories were assessed using the Wilcoxon rank sum and Kruskal-Wallis tests, respectively. Age and gender were not significantly related to bias, but PI differed between women (median [IQR]: 2.0 [0.8- 3.5]) and men (4.5 [2.0-7.4]), P<0.0001. Perfusion index did not vary between light, medium and dark skin categories, P=0.85. The distribution of skin pigmentation categories did not differ between men and women, P=0.41

[0096] Receiver operator characteristic (ROC) curves were constructed for the definition of hypoxemia (SaO2 < 88%) at various SpO2 thresholds for the Masimo and Nellcor pulse oximeters.

“ Missed hypoxemia” was defined as the percentage of pulse oximeter readings with SpO2 at or above cutoff values of 92%, 94% and 96% with corresponding SaO2< 88%.

[0097] Results

[0098] Study population and sample size.

[0099] Data was used from 146 consecutive, healthy, nonsmoking volunteer participants in pulse oximeter performance studies at the University of California at San Francisco (UCSF) Hypoxia Research Laboratory in 2020 and 2021. Demographic information on the study population is presented in FIG. 7. Forty-three subjects classified as Fitzpatrick skin type V or VI (Dark), 78 as Fitzpatrick III or IV (Medium) and 25 as Fitzpatrick I or II (Light). A total of 9,763 arterial blood gas samples were analyzed with 1,774 from Fitzpatrick I- II, 5,133 from Fitzpatrick III-IV, and 2,856 samples from Fitzpatrick V-VI volunteers (FIG. 7). Each subject participated in an average (SD) of 2.8 o } 3.0 studies (median [IQR]: 1 [1-3]; range 1-16) studies with 65 subjects who had data from >1 study.

[00100] Primary Outcomes.

[00101] A multivariable analysis accounting for repeated measures revealed that skin pigment, perfusion index and degree of hypoxemia significantly contributed to the amount of error (bias) in both pulse oximeters (FIG. 8) as significant interaction was found between these three variables. FIGS. 9A-C) summarize bias statistics for each of the variables separately. For both pulse oximeters, mean bias and Arms increased with skin pigmentation class, and increased with decreasing perfusion. Since gender was highly correlated to PI, the model was also run separately for men and women with the same result (data not shown). The interaction between hypoxemia and PI in darkly pigmented subjects shows that positive bias increases synergistically during hypoxemia with lower perfusion, an effect seen to a smaller extent in medium and lightly pigmented subjects (FIGS. 10A-B). The shift in distribution of pulse oximeter readings towards larger errors during low perfusion differed between subjects with light, intermediate and dark skin (FIGS. 11A-B).

[00102] Errors in pulse oximetry and missed diagnosis of hypoxemia.

[00103] Pulse oximeter errors near the threshold for the diagnosis of hypoxemia (SaO2 88-90%) have special clinical importance. In FIG. 12, pulse oximeter readings and corresponding SaO2 were compared, identifying values for which the pulse oximeter read 92% or greater when SaO2 was <88%, representing a missed hypoxemia diagnosis. Missed hypoxemia was more frequent with darker skin and lower perfusion for both oximeters tested. For subjects with darkly pigmented skin and low perfusion, missed hypoxemia for the SpC range of 92-96%was seen in 30.2% of the readings from the Masimo pulse oximeter and 7.9% of the readings from the Nellcor. The same trend was seen when different thresholds for errors were analyzed. The frequency of missed hypoxemia for SpO2 thresholds of >92%, >94%, >96% and 92-96% are summarized in FIG. 13. [00104] The area under the receiver operating characteristic curve (ROC) for a diagnosis of hypoxemia was lowest for subjects with darkly pigmented skin and low perfusion, 0.96 (95% CI: 0.95-0.98). Sensitivity cutoff values were shifted to higher perfusion values in subjects with dark skin. For example, to achieve 95% sensitivity for SaO2 < 88% for the Masimo in darkly pigmented subjects with PI<1%, a cutoff of = 95% would be needed, whereas this cutoff would correspond to 99% sensitivity for SaO2 < 88% in darkly pigmented subjects with PI ^2% (data not shown).

[00105] Discussion

[00106] In this prospective study of healthy volunteers, it was found that low perfusion interacts with both medium and dark skin pigment to substantially increase pulse oximeter bias. [00107] Pulse oximeter bias and missed hypoxemia diagnosis.

[00108] Missed diagnosis of hypoxemia occurred at higher rates both for subjects with not only dark skin pigment but also medium skin pigment. The combination of dark skin and low perfusion causes errors in pulse oximeter readings large enough to miss the diagnosis of hypoxemia in 8- 30% of readings when SaO2 is <88% and the pulse oximeter reads 92-96% (FIG. 14). In darkly pigmented subjects across all ranges of perfusion, the missed diagnosis rate was about 3-9%, and for subjects with medium skin pigmentation, the missed diagnosis rate was about 1-2%. Subjects with medium skin pigment in the study self-identified as Asian, Hispanic, or Multiethnic. The findings remain robust when different ranges of hypoxemia are used as the threshold for a finding of missed diagnosis (FIG. 14).

[00109] Low Perfusion and clinically significant pulse oximeter errors.

[00110] The thresholds for low perfusion were selected based on a sensitivity analysis of the relationship of PI to level of pulse oximeter bias. In clinical experience in the ICU and perioperative settings, low peripheral perfusion was commonly encountered due to underlying disease as well as lack of routine warming of patient hands during oximetry monitoring. Poor peripheral perfusion is common in the elderly and in patients with sepsis, peripheral vascular disease, hypertension, diabetes, and other conditions (Falotico IM, Shinozaki K, Saeki K, Becker LB. Advances in the Approaches Using Peripheral Perfusion for Monitoring Hemodynamic Status. Front Med (Lausanne). 2020;7:614326.)

[00111] Conclusions

[00112] Low perfusion is associated with misdiagnosis of arterial hypoxemia by pulse oximetry in healthy subjects with dark skin pigmentation in controlled laboratory conditions. Significant pulse oximeter bias occurring in subjects with medium and dark skin pigment and low perfusion was obscured when overall performance (i.e., all subjects) of the pulse oximeters was analyzed (FIG. 15). Both instruments performed within the 3% specification for FDA clearance.

[00113] The physiological control with stable levels of hypoxemia made possible by this study design produced better synchronization of stable SpO2 and measured SaO2 than is possible in retrospective clinical studies. Moreover, multiple data points for different individuals under different conditions allowed for an extremely robust analysis of the impacts of perfusion on bias. Also, the study involved two pulse oximeter models widely used in the clinical setting;

[00114] Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, it is readily apparent to those of ordinary skill in the art in light of the teachings of this invention that certain changes and modifications may be made thereto without departing from the spirit or scope of the appended claims.

[00115] Accordingly, the preceding merely illustrates the principles of the invention. It will be appreciated that those skilled in the art will be able to devise various arrangements which, although not explicitly described or shown herein, embody the principles of the invention and are included within its spirit and scope. Furthermore, all examples and conditional language recited herein are principally intended to aid the reader in understanding the principles of the invention and the concepts contributed by the inventors to furthering the art, and are to be construed as being without limitation to such specifically recited examples and conditions. Moreover, all statements herein reciting principles, aspects, and embodiments of the invention as well as specific examples thereof, are intended to encompass both structural and functional equivalents thereof. Additionally, it is intended that such equivalents include both currently known equivalents and equivalents developed in the future, i.e., any elements developed that perform the same function, regardless of structure. Moreover, nothing disclosed herein is intended to be dedicated to the public regardless of whether such disclosure is explicitly recited in the claims.

[00116] The scope of the present invention, therefore, is not intended to be limited to the exemplary embodiments shown and described herein. Rather, the scope and spirit of present invention is embodied by the appended claims. In the claims, 35 U.S.C. § 112(f) is expressly defined as being invoked for a limitation in the claim only when the exact phrase "means for" or the exact phrase "step for" is recited at the beginning of such limitation in the claim; if such exact phrase is not used in a limitation in the claim, then 35 U.S.C. § 112(f) is not invoked.

Claims

CLAIMS What Is Claimed Is:

1. A method of assessing blood oxygenation of a subject with a pulse oximeter and treating the subject accordingly, the method comprising: recording data from a pulse oximeter positioned on a body part of the subject, wherein the pulse oximeter data comprises the amount of light emitted from the pulse oximeter that is transmitted through the body part over time; determining a raw peripheral oxygen saturation (raw SpC ) and a perfusion index based on the pulse oximeter data; determining skin pigmentation of the subject; adjusting the raw SpCh based on the perfusion index and the skin pigmentation of the subject, thereby generating an adjusted SpCh and a confidence interval for the adjusted SpCh; determining a high probability of hypoxemia based on the adjusted SpCh, the confidence interval, or a combination thereof; and treating the subject for the high probability of hypoxemia.

2. The method of claim 1, wherein the adjusting is performed while considering the perfusion index and the skin pigmentation together.

3. The method of claim 1, wherein the adjusting based on the perfusion index is performed independently from the adjusting based on the skin pigmentation.

4. The method of any one of claims 1-3, wherein the skin pigmentation is determined from the pulse oximeter data.

5. The method of claim 4, wherein determining skin pigmentation from the pulse oximeter data comprises comparing an amount of visible light transmitted through the body part with an amount of infrared light transmitted through the body part.

6. The method of claim 5, wherein the amount of visible light transmitted through the body part is red light.

7. The method of any one of claims 1-3, wherein the skin pigmentation is determined by visual observation of the subject or from a medical database.

8. The method of any one of claims 1-7, wherein the perfusion index is the ratio between AC and DC, wherein AC is peak-to-peak amplitude of the transmitted light over time, wherein DC is the lowest light transmittance value.

9. The method of any one of claims 1-8, wherein the raw SpCh is adjusted based on historical data from multiple individuals comprising a comparison of peripheral oxygen saturation (SpCh), arterial oxygen saturation (SaCh), skin pigmentation, and perfusion index.

10. The method of any one of claims 1-9, wherein the confidence interval is a 95% confidence interval.

11. The method of any one of claims 1-10, wherein determining the high probability of hypoxemia is based on a low adjusted SpC .

12. The method of any one of claims 1-11, wherein determining the high probability of hypoxemia is based on a large confidence interval.

13. The method of any one of claims 1-12, wherein treating the subject comprises measuring arterial oxygen saturation (SaCh) with an arterial blood gas test.

14. The method of any one of claims 1-13, wherein treating the subject comprises administering oxygen-enriched air to the subject, administering a bronchodilator to the subject, administering a diuretic to the subject, or a combination thereof.

15. The method of any one of claims 1-14, wherein the body part is a finger or an ear lobe.

16. A pulse oximeter for assessing blood oxygenation in a subject and communicating the results of the assessment, wherein the pulse oximeter is configured to: record pulse oximeter data from a body part of the subject, wherein the pulse oximeter data comprises the amount of fight emitted from the pulse oximeter that is transmitted through the body part over time; determine a raw peripheral oxygen saturation (raw SpO?) and a perfusion index based on the pulse oximeter data; determine skin pigmentation of the subject based on the pulse oximeter data or receive skin pigmentation data; adjust the raw SpCh based on the perfusion index and the skin pigmentation of the subject, thereby generating an adjusted SpOz and a confidence interval for the adjusted SpCh; and communicate the adjusted SpC and confidence interval.

17. The pulse oximeter of claim 16, wherein the communicating comprises providing a visual indication of the adjusted SpCh and confidence interval.

18. The pulse oximeter of claim 17, wherein the visual indication is provided by a visual display of the pulse oximeter.

19. The pulse oximeter of claim 17, wherein the visual indication is provided by a visual display that is separate from the pulse oximeter.

20. The pulse oximeter of any one of claims 16-19, wherein the communicating comprises providing an auditory alert that the adjusted SpO: or confidence interval is outside of a predetermined range.

21. The pulse oximeter of any one of claims 16-20, wherein the confidence interval is a 95% confidence interval.

22. The pulse oximeter of any one of claims 16-21, wherein the adjusting is performed while considering the perfusion index and the skin pigmentation together.

23. The pulse oximeter of any one of claims 16-21, wherein the adjusting based on the perfusion index is performed independently from the adjusting based on the skin pigmentation.

24. The pulse oximeter of any one of claims 16-23, wherein the skin pigmentation is determined from the pulse oximeter data.

25. The pulse oximeter of claim 24, wherein determining skin pigmentation from the pulse oximeter data comprises comparing an amount of visible light transmitted through the body part with an amount of infrared light transmitted through the body part.

26. The pulse oximeter of claim 25, wherein the amount of visible light transmitted through the body part is red light.

27. The pulse oximeter of any one of claims 16-23, wherein the pulse oximeter is configured to receive skin pigmentation data.

28. The pulse oximeter of any one of claims 16-27, wherein the perfusion index is the ratio between AC and DC, wherein AC is peak-to-peak amplitude of the transmitted light over time, wherein DC is the lowest light transmittance value.

29. The pulse oximeter of any one of claims 16-28, wherein the raw SpC is adjusted based on historical data from multiple individuals comprising a comparison of peripheral oxygen saturation (SpCh), arterial oxygen saturation (SaCh), skin pigmentation, and perfusion index.

30. A kit for assessing blood oxygenation in a subject and communicating the results of the assessment, the system comprising: a pulse oximeter of any one of claims 16-29; and packaging containing the pulse oximeter.

31. A method of determining skin pigmentation of a body part of a subject, the method comprising: recording an amount of visible light emitted from a light source that is transmitted through the body part; recording an amount of infrared light emitted from the light source that is transmitted through the body part; and comparing the ratio of transmitted visible light to transmitted infrared light and thereby determining skin pigmentation at the body part.

32. The method of claim 31 , wherein the visible light is red light.

33. The method of claim 32, wherein the red light comprises light at 660 nm.

34. The method of any one of claims 31-33, wherein the infrared light comprises light at 900 nm.

35. The method of any one of claims 31-34, wherein the light source is a pulse oximeter.

36. The method of any one of claims 31-35, wherein the body part is a finger.

37. The method of any one of claims 31-35, wherein the body part is an ear lobe.

38. A controller for determining skin pigmentation of a body part of a subject, wherein the controller is configured to: record an amount of visible light emitted from a light source that is transmitted through the body part; record an amount of infrared light emitted from the light source that is transmitted through the body part; and compare the ratio of transmitted visible light to transmitted infrared light and thereby determining skin pigmentation at the body part.

39. The controller of claim 38, wherein the visible light is red light.

40. The controller of claim 39, wherein the red light comprises light at 660 nm.

41. The controller of any one of claims 38-40, wherein the infrared light comprises light at 900 nm.

42. The controller of any one of claims 38-41, wherein the light source is a pulse oximeter.